Estimating of residual load-carrying capacity of CFST stub columns after exposure to elevated temperatures using gene expression programming

Nowadays, the use of composite sections has become a common practice in the construction industry. Concrete is inherently a brittle material, with high stiffness and compressive strength. On the other hand, steel is a material with high tensile strength and ductility. The simultaneous use of steel and concrete in composite sections improves the performance and leads to optimum exploitation of the properties of both steel and concrete materials. Concrete-filled steel tube (CFST) is a type of section often used in high-rise buildings. In addition, the composite action of steel and concrete in CFST columns gives some advantages to these sections during fire incidents. On the one hand, the concrete core prevents the local buckling of the steel tube, and on the other, the steel tube prevents the spalling of concrete at elevated temperatures. The behavior of CFST sections at elevated temperatures is complicated due to interactions between the steel tube and concrete core. Therefore, achieving a correct understanding of the behavior and material properties in CFST columns is required for design and strengthening purposes. In this research, with the help of the gene expression programming (GEP) technique, a formula was developed to estimate the ultimate load-carrying capacity of CFST columns after exposure to elevated temperatures. To that end, the experimental data of 94 groups of CFST stub columns were employed, of which 80% were used to train the model and the remaining 20% to validate the model. Input variables included the compressive strength of the concrete core ( ), cross-sectional area of the concrete core ( ), yielding stress of steel ( ), cross-sectional area of steel tube ( ), normalized temperature ( ), and the confinement index ( ). The validity of the developed model was assessed using a portion of the data that had not been employed in the training phase. To ensure the correct prediction of the ultimate load-carrying capacity of CFST stub columns by the developed model, a sensitivity analysis and parametric studies were conducted on the model and revealed the complete compatibility of the model with physical facts. The results of this research indicate that increasing the compressive strength of the concrete core, cross-sectional area of the steel tube, yield stress of steel tube, cross-sectional area of the concrete core and the confinement index increases the ultimate load-carrying capacity of the CFST section, while increasing the exposure temperature lowers this parameter.